Back pain is one of the most frequently reported musculoskeletal problems in the United States, commonly caused by degeneration of a disc. The vertebrae of the spine articulate with each other to allow motion in the frontal, sagittal and transverse planes, but as the weight of the upper body increases, the vertebral bodies, which are designed to sustain mainly compressive loads, increase in size caudally. The intervertebral discs, comprising about 20-33% of the lumbar spine length, is a major link between the adjacent vertebrae of the spine, which with the surrounding ligaments and muscles, provide stability to the spine. Each disc consists of a gelatinous nucleus pulposus surrounded by a laminated, fibrous annulus fibrosus, situated between the end plates of the adjacent vertebrae. The discs sustain weight and transfer the load from one vertebral body to the next, and maintain a deformable space to accommodate normal spine movement.
The nucleus pulposus contains collagen fibrils and water-binding glycosaminoglycans, while the surrounding annulus fibrosus comprises fibrocartilaginous tissue and fibrous protein arranged in 10 to 20 lamellae forming concentric rings around the nucleus pulposus. The collagen fibers within each lamella are parallel to each other, angled at ˜60° from vertical, and alternating in direction of the inclination for each lamellae. This crisscross arrangement enables the annulus fibrosus to withstand torsional and bending loads. The end-plates comprise hyaline cartilage, and are directly connected to the lamellae, forming the inner one-third of the annulus.
Under compressive loads, the nucleus pulposus flattens and bulges radially outward; while the annulus fibrosus stretches, resisting the stress. The end-plates of the vertebral body also resist the ability of the nucleus pulposus to deform. Thus, pressure is applied against the annulus and the end-plate, transmitting the compressive loads to the vertebral body. Conversely, when tensile forces are applied, the disc is raised, straining the collagen fibers in the annulus. When the individual bends, one side of the disc is in tension while the other side is in compression, and the annulus on the compressed side bulges outward.
When the disc is subjected to torsion, shear stresses are applied. As an individual ages, repeated rotational loading initiates circumferential tears in the annulus fibrosus, which gradually causes radial tears in the nucleus pulposus, resulting in degradation and water loss at the rupture. Consequently, over time a disc looses its ability to resist compressive loads as the annulus bulges. As the severity of the tear increases, much of the disc contents are lost, leaving only a thin line of fibrous tissue at the nuceous pulposus—a condition known as disc resorption. As the annulus bulges posteriorly into the spinal canal, compression of the nerve root results in sciatica. Once the disc ruptures, excessive motions can cause in spine segmental instability, making the spine more vulnerable to trauma and degenerative spondylolisthesis and increasing pain.
Surgical treatments for herniated disc include laminectomy, spinal fusion and disc replacement with prostheses. However, the results of spinal fusions are varied, even with autografts, and result in total elimination of motion at the fused joint and often significant long-term limitations. Therefore, as an alternative to spinal fusion, intervertebral disc prostheses have been used to replace and simulate the function of a normal disc. Disc replacement prostheses must be biocompatible and capable of sustaining weight and transferring load from one vertebral body to the next. Such a prosthesis should last for the lifetime of the patient and have sufficient mechanical strength to resist injury during movement, while maintaining a deformable space between the vertebrae to accommodate movement. More importantly, the replacement disc should be able to be fixed to the adjacent vertebrae, but should not cause damage if the disc fails.
Many artificial disc designs have been used or proposed. Devices for total disc replacement have ranged from metal ball bearings with a silicone rubber nucleus and silicone fluid-filled plastic tube, to spring systems, elastic disc prosthesis, and elastic discs encased in a rigid column. Unfortunately, over time such devices have proven to offer inadequate mechanical performance and unlikely long-term bone fixation.
However, replacing only the nucleus prosthesis has several important advantages over total disc replacement. The nucleus has a much simpler structure and function than the annulus and endplates, so that with proper design of the nucleus prosthesis the surgeon can leave the annulus and endplates intact. Only the spatial and mechanical properties of the annulus are replaced to relieve the compressive load on the disc. This permits the remaining disc tissues and their functions to be retained, while requiring only a minimally invasive surgical procedure, such as endoscopic implantation through a small incision in the annulus. The implant can be designed so that no fixation to the vertebra is required.
Various attempts have been made to develop prostheses for the nucleus, using for example, metal (Fernstrom, Nord. Med. 71:160 (1964)), or various elastomers (either preformed or formed in situ) (Hou et al., Chinese Med. J. 104:381-386 (1991), but proved to lack biocompatibility and mechanical strength. Bao et al. designed a much more effective artificial intervertebral disc nucleus (Bao et al., Biomaterials 17:1157-1167 (1996); U.S. Pat. Nos. 5,047,055 and 5,192,326) using hydrogel materials containing approximately 70% water content under physiological loading conditions. The hydrogel nucleus implant mimics the capacity of the nucleus tissue to absorb and release water under cyclic loading conditions, thereby mechanically restoring disc anatomy and function, without adverse local or systemic tissue reaction. Hydrogels have several major mechanical drawbacks. By definition, they operate by absorbing water after cross linking, but for example, in the degenerated disc site, water absorption cannot be controlled. Moreover, water absorption is a reversible process, i.e., the water is also lost from a hydrogel implant, as well as absorbed—which again renders control of hydration very difficult. In addition, the thermal activation procedures necessary to achieve crosslinking of some of the hydrogels are notoriously difficult to control.
An interbody rigid, threaded nucleus implant has been clinically used, see Ray et al., The Artificial Disc, (Brock et al., eds), Berlin: Springer-Verlag, 1991, pp 53-67; U.S. Pat. Nos. 4,772,287, 4,904,260 and 5,674,295. The prostheses have a hydrogel core constrained by a flexible, but inelastic woven polyethylene-fiber jacket intermingled with bioresorbable materials to attract tissue ingrowth, but implantation requires discectomy using a larger-than normal incision. However, the prosthesis materials are purely synthetic, including no cells or developing tissue. More recently, to minimize implant extrusion and minimize invasive surgery Advanced BioSurfaces has developed a system featuring a polymer formulation delivered through a balloon catheter using a polymer-injection gun (U.S. Pat. No. 5,888,220). Polyurethane is used and cured in situ to provide strong mechanical properties. However, many of the components of the system lack biocompatibility, and the use of a balloon catheter leads to a more involved treatment procedure and mechanical limitations upon treatment.
For repairing skin tissue, particularly of burn patients or for decubitus wounds, cells have been seeded onto templates of either resorbable or non-resorbable material and dressed onto the wound site (Sabolinski, Biomaterials 17:311-320 (1996)). However, tissue engineering of the skin is significantly different from tissue engineering of the intervertebral disc, requiring completely different tissue compositions and unique mechanical requirements.
U.S. Pat. Nos. 5,108,438 and 5,258,043 teach a porous matrix of biocompatible and bioresorbable fibers which may be interspersed with glycosaminoglycan molecules to form a scaffold matrix for regenerating disc tissue and replace both the annulus fibrosus and nucleus pulposus. However, replacement of so much tissue is a relatively invasive procedure, requiring lengthy recovery time. Furthermore, these matrices include no cells to stimulate tissue recovery, nor is there any incipient tissue formation in this device at the time of implantation. Polymeric scaffolds for tissue engineering are well known (e.g., Thomson, “Polymer Scaffold Processing,” pp. 251-261 in Principles of Tissue Engineering, 2nd Ed., (Lanza et al., eds.) Academic Press, San Diego, Calif., 2000). However, use of these pre-made structures require extensive surgery.
Various bioactive glass materials (e.g., U.S. Pat. No. 5,204,104) have been seeded with cells in order to facilitate cell function including proliferation and extracellular matrix synthesis, e.g., El-Ghannam, et al., J. Biomed. Materials Res. 29:359-370 (1974), Schepers et al., J. Oral Rehab. 18:439-452 (1991) and U.S. Pat. No. 5,204,106, have been used. In addition, dense bioactive glass discs have been found to enhance osteoprogenitor cell differentiation (Baldick et al., Transactions 5th World Biomaterials Conf., Toronto, II-114 (June 1996), and have been seeded with cells in order to facilitate cell function including proliferation and extracellular matrix synthesis, e.g., El-Ghannam, et al., J. Biomed. Materials Res. 29:359-370 (1974), Schepers et al., J. Oral Rehab. 18:439-452 (1991) and U.S. Pat. No. 5,204,106.
U.S. Pat. No. 5,964,807 teaches a method for evacuating degenerated nucleus pulposus tissue from the intervertebral disc, then implanting into the evacuated nucleus pulposus space, a hybrid material prepared by combining intervertebral disc cells with a biodegradable substrate comprising bioactive glass, polymer foam, or polymer foam coated with sol gel bioactive material. The polymer foam comprises copolymers of D,L poly(lactide-co-glycolide). The process includes isolating viable cells from the nucleus pulposus of an intervertebral disc and culturing the cells on the biodegradable substrate prior to implantation, or culturing the cells prior to combining them with the biodegradable substrate immediately prior to implantation. But the resulting implant lacks the required viscoelastic properties of the healthy nucleus pulposus, e.g., latridis et al., J. Orthopaedic Res. 17:732-737 (1999); latridis et al., J. Orthopaedic Res. 15:318-322 (1997).
Liquid and foam emulsions have been described for the preparation of polymeric scaffolds for tissue engineering; e.g., U.S. Pat. Nos. 6,337,198 and 6,103,255 teach the preparation of porous polymer scaffolds from an emulsion of a selected polymer and multiple solvents, then driving off the solvents to form pores, thereby creating a scaffold. These approaches require that first the polymer is formed, and then it is emulsified. However, polymerization of the scaffold is thermal-based, making polymerization in situ in the patient unacceptable there may be residual organic solvents that could cause inflammatory or toxic reactions.
Water-in-oil (inverse) (“W/O”) emulsions consisting of a discrete aqueous phase and an organic, typically hydrocarbon, continuous phase have diverse uses in coatings, adhesives, water treatment, cosmetics, personal care and pharmaceuticals. W/O emulsions are valuable largely because they supply high concentrations of water-soluble “active agents” in a liquid, free-flowing form. See, e.g., Friberg, et al., “Emulsions,” in Kirk Othlner Encyclopedia of Chemical Technology, 4th Ed., Vol. 9, pp.393-413, 1994. Such active agents may include one or more components, e.g., high molecular weight, water soluble polymers for water treatment, or organic salts, e.g., many drugs for treatment of disease.
Processes for the polymerization of reactable monomers and oligomers are well know in the art, and are commonly used industrially for preparing coatings, inks, and other polymeric systems. See, e.g., generally, Allcock et al., Contemporary Polymer Chemistry, 2nd Ed., Prentice Hall, Englewood Cliffs, N.J., 1990, p. 553. Often light, particularly UV light, is used to initiate the polymerization process, e.g., photopolymerization.
EP 0430 517(A2) teaches biomosaic polymers formed from W/O emulsions and microemulsions, used to produce membranes for assays, separations, and catalysis processes that require the binding of biologically active materials, e.g., proteins, enzymes, DNA, antigens and antibodies, at the polymerized surfaces. The exemplified monomers are isobornyl acrylate and methyl methacrylate, and binding of the active materials may be chemical or physical, depending on the intended use. The methods taught include photoinitiated polymerization of the surfactant-stabilized emulsions. Similarly, U.S. Pat. No. 5,151, 217 discloses bicontinuous microemulsions in which both the organic (oil) phase and the aqueous (water) phase contain polymerizable organic monomers. The aqueous phase also contains a polymerizable surfactant, and apparently polymerization may be activated by the inclusion of photoinitiators. Conversely, oil-in-water (“O/W”) emulsions, wherein water is the continuous phase, representing a structural mechanical component of the final product, polymerization has also been photoinitiated, e.g., generally, Costanza et al., Radiation Cured Coatings, Federation of Societies for Coating Technology, Philadelphia, Pa., 1986; U.S. Pat. No. 6,045,972.
Polymers also can be prepared in gel form. The polymer gel is essentially a crosslinked solution, linked by chemical bonds, crystallization or other kinds of junctions made by means of the chemical properties of the monomers within the polymer. “Hydrogels” are a specific class of polymer gels, comprising crosslinked polymer networks of hydrophilic, water soluble or water dispersible polymers, including polyelectrolytes, which are permeated by, and swell in, aqueous media—but do not dissolve therein (Ferry, Viscoelastic Properties of Polymers, John Wiley, NY, 1980, p.529; Allcock et al., supra; DeRossi et al., in Macro-Ion Characterization, (Smith, ed.), ACS Symposium Series 548, Amer. Chem. Soc., Washington, D.C., 1994, p. 20).
Injectable hydrogels, gelatins and similar polymeric carriers for tissue engineering applications are well known in the art. For example, in U.S. Pat. No. 6,224,893, Langer et al. teach the use of interpenetrating polymer networks that form hydrogels for drug delivery and tissue engineering, reportedly useful for injectable, photocurable cell scaffold formation. However, these interpenetrating networks and the precursor components comprise single-phase dispersion requiring two or more polymers, the crosslinking of which results in a continuous hydrogel state with limited mechanical properties.
Ideally, intervertebral disc treatment would guide and stimulate reformation of damaged or diseased intervertebral disc tissue, especially nucleus pulposus and annulus fibrosus tissue. Currently, however, although hydrogel and tissue engineering models have been tried, there is still not a good therapeutic approach for reconstructing damaged discs, using non-toxic, biocompatible replacement materials that will not elicit an autoimmune or inflammatory response that might result in rejection. Such materials should be deliverable in a fluid or gel state by minimally invasive methods, but which when set, should fill the entire internal space of the intervertebral disc to provide mechanical properties comparable to the nucleus pulposus. Thus, a need for such a material has remained until the present invention.